Abuse liability assessment in preclinical drug development: Predictivity of a translational approach for abuse liability testing using methylphenidate in four standardized preclinical study models

Abstract

Objectives: Preclinical abuse liability assessment of novel clinical CNS-active candidates involves several tests, addressing different aspects characteristic for abuse potential, which are considered predictive for substance abuse of these candidates, thus ensuring an appropriate translational approach. To demonstrate how such a strategy could work, a known drug of abuse, methylphenidate was evaluated in a full rodent test battery, comprising four test models, and in accordance with the requirements of the FDA, ICH and EMA guidelines. Methods: Methylphenidate was tested orally at 2.5, 5 or 10 mg/kg for its physical dependence potential in a repeated dose non-precipitated withdrawal test, for its drug profiling in a drug discrimination learning procedure (single escalating doses), and for its reinforcing properties in a conditioned place preference test (alternate dosing days) and an intravenous self-administration procedure (0.05 to 1 mg/kg/IV infusion during 5 daily 1-h test sessions). The stimulant D-amphetamine served as positive control and was administered subcutaneously at 0.8 mg/kg in the first three test models. In the intravenous self-administration procedure rats were habituated to intravenously self-administer D-amphetamine at 0.06 mg/kg/IV infusion prior to methylphenidate substitution. Results: Cessation of subchronic dosing up to 10 mg/kg methylphenidate led to sustained or even exacerbated effects on locomotion and behavior, body temperature, body weight, food consumption, and alteration of the diurnal rhythm during withdrawal. Clear generalization to D-amphetamine was obtained in the drug discrimination test at 5 and 10 mg/kg. Distinct reinforcing properties were present in the conditioned place preference test at 10 mg/kg and in the intravenous self-administration study from 0.05 mg/kg/IV infusion onwards. The maximum plasma exposure after oral administration of methylphenidate over the dose ranges tested in the present rat studies covered at least 1.9-fold to 18.9-fold the recommended human therapeutic exposure of 10 ng/ml, a plasma level that is considered representative of the human efficacious methylphenidate dose. The ratio C-max Hu/rat calculated from the intravenous self-administration data ranged from 14.9 to 576.5. Consequently the regulatory requirements, stating that preclinical drug abuse liability studies should include high doses that produce plasma levels that are multiples of the therapeutic dose were fulfilled (FDA, EMA, ICH). Discussion: The presented preclinical models, implemented within a drug development environment, were considered highly predictive to assess the abuse potential of methylphenidate, and in accordance with the regulatory requirements of drug licensing authorities in terms of appropriate methods, dose selection and subsequent plasma exposure. (C) 2014 Elsevier Inc. All rights reserved.